Once upon a time, a new territory was discovered. It was a fertile land, promising opportunity for settlers who had struggled in their old countries. As the colonists arrived – willing to endure frontier conditions because of a belief in what-might-be – leaders stepped forward with good suggestions, bright ideas, and plain old management skills. Before long, there were several such leaders – and they began to fight over the territory.
This is not a retelling of a mythic time, nor is it a new science fantasy series. It’s the history of nearly every technology adoption. First, there’s a discovery. Then we have a chaotic period of innovation, when proto-leaders put forward good ideas, or at least propose things worth trying out. It’s followed by an era of consolidation, during which some initial innovators fall away, due to technical failures or business weakness. Those initial innovators may be acquired (or conquered) by those who remain. In a relatively short period of time, only a few leaders rule the lands – or Kingdoms.
And we settlers, who live in that territory – or who use that technology – need to be aware of the wars between the Kingdoms. Because it’s we who fight in those wars, who choose (and change) alliances, who build our homes in that territory.
Early this year, I wrote about the trends I saw coming in the container space, Winter is Coming. I wrote, “It’s clear that the market in which software containers like Docker, Kubernetes, and Mesos operate is shifting. I see several things changing in the container ecosystem in 2017 – and they’re probably not what you’re thinking.” I suggested that this would be the year that containers sorted themselves out.
I thought with conference season almost behind us (except KubeCon + CloudNativeCon North America of course) I’d take a moment and reflect what’s happened in the last 10 months.
If the Land of the Container Ecosystem was ruled by four Kingdoms – Docker, Kubernetes, Mesos, and Cloud Foundry – I was sure that there would be a war among them. It would come to blows pretty soon, I predicted. However, I thought that by now the war would be over, with only a single Kingdom left to rule the container space. But that isn’t quite how it played out.
I feel as though I could write an entire novel about these Kingdoms: Mesos, who was an early settler in the territory, who built the first towns, who ruled fairly for many years. Cloud Foundry who also was an early settler with a devout following. Kubernetes, the shiny young upstart who has earned the people’s love. And Docker, who was most clever… until it stumbled.
And, surely, Docker stumbled – at the Battle of DockerCon US. DockerCon was held in April. It was a perfect opportunity for the company to shine: Docker had all the spotlight and held the mindshare. The company could prove to the land that they could unite the kingdom with a brilliant product strategy. Instead, Docker launched Moby and confused everyone with the company’s business vision. Personally, I walked away from that event and thought, “They’re done. I don’t know what their viable business is.” I don’t think I was the only one. The company leadership spent weeks issuing clarifications and doing damage control, that only served to muddy the waters.
The Battle of DockerCon Europe saw perhaps a resurgence of their relevance when they announced bending the knee to Kubernetes. Was it a shot of brilliance, or simply more evidence of lack of a clear strategy? It seems these days when you’re in this market and searching for relevance, you hitch your wagon to Kubernetes.
My prediction is that Docker will be sold. My first thought of a suitor is Microsoft. Why Microsoft? It’s an easy fit. Microsoft knows how to work with large organizations. Among that wisdom is the understanding that enterprises are slow to make technology moves. Buying Docker would be a way to support those enterprise customers with a ready-to-use and accepted container brand. That isn’t a bad scenario. Microsoft is becoming more active in the open source space – and taking it seriously. Microsoft recently hired a few well known open source developers from Google.
In my view, Kubernetes has won the mindshare of the community (or populace). The entire industry has rallied behind Kubernetes even though it’s less mature (though making tremendous strides) and not quite as battle tested. Kubernetes was donated to Cloud Native Computing Foundation (CNCF) in the 1.0 release timeframe (2015) and has become one of the anchor projects. One thing that Kubernetes does really well is build community. This certainly helps because the community is so well funded, both directly by Google and through other parties paying to support it through the CNCF. That aside, Kubernetes has exemplified transparency and inclusion through its use of special interest groups. These Special Interest Groups allow a large group of people to feel included and heard, and are made up with members from different companies and organizations, all with a common interest in a specific topic. These SIGs help nurture and distribute leadership, while advancing new proposals, designs and release updates.
Kubernetes rose to power because it’s good, kind, and (let’s not forget) powerful benefactor Google funded it. Can Kubernetes continue to rule if Google does not invest so much money and effort? Google has been known to pull the plug on projects that do not substantially benefit its cloud or other strategic directions. Can Kubernetes survive if its benefactor does not support it anymore? I don’t know, but I don’t think it is wise to underestimate it. It is, after all, the fastest growing open source project since the Linux Kernel.
Then we have Mesos. It’s no secret at this point that while at Apple I was part of the team that moved the Siri infrastructure to Mesos. So I do have a soft spot for it. I still believe that its two-tier application aware architecture makes it incredibly powerful. It seems then, for the time being, that others that have settled in the container kingdom have realized this power and voted with their wallets. Mesosphere recently announced just how well they’ve done in this space. Though, like Docker, Mesosphere has also bent the knee to Kubernetes with its announcement of an initiative to support Kubernetes on DC/OS.
Finally there is Cloud Foundry. CF built its kingdom among the Fortune 100 companies with a highly powerful, and cleverly crafted, product offering. Cloud Foundry is also supported by the Cloud Foundry Foundation, which boasts 65 contributing members. Cloud Foundry realized that in order to provide a meaningful platform, simply providing a runtime solution wasn’t enough. Instead, Cloud Foundry ties into the software development process and includes tooling around continuous integration and deployment workflows. CF has been wildly successful on delivering a powerful experience to developers in part in efforts to retrain entire software development teams. This has left Pivotal (the main company behind CF) with a devout follower base, and a war chest last year that brought in $270M. Then even the mighty Cloud Foundry, following in its competitors footsteps, also bent the knee to Kubernetes with its PKS project.
Like many people, I tend to think in terms of “there can be only one” winner – but that isn’t necessarily so. Some argue that the marketplace in general permits two dominant vendors (Hertz versus Avis) and everyone else is an also-ran. Alternatively, the “Kingdoms” may still be jockeying for position, and more battles will come, for now though, winter has come.
Now: What happens next? Here are some things worth contemplating:
- Is there room for two Kingdoms in the Land of the Container Ecosystem? Even if the second hasn’t garnered significant support and allies? If the company makes enough money, perhaps it doesn’t matter.
- Will Kubernetes survive if Google isn’t happy with its progress? Are they content to win on notoriety if there is no impact to their checkbook? Is there proof investment in Kubernetes driving Google Cloud adoption?
- What happens to Docker the company, its technology that is used-and-appreciated, and the passionate community that supports it?
Stay tuned as we watch the battle for the iron throne continue to rage throughout 2018.